✨ Share maven cache between analyze and dep, use JVM max setting

cmd/analyze.go

@@ -28,6 +28,8 @@ import (
 )
 
 var (
+	// TODO (pgaikwad): this assumes that the $USER in container is always root, it may not be the case in future
+	M2Dir = path.Join("/", "root", ".m2")
 	// application source path inside the container
 	SourceMountPath = path.Join(InputPath, "source")
 	// analyzer config files
@@ -55,6 +57,7 @@ type analyzeCommand struct {
 	output                string
 	mode                  string
 	rules                 []string
+	jaegerEndpoint        string
 
 	// tempDirs list of temporary dirs created, used for cleanup
 	tempDirs []string
@@ -151,6 +154,7 @@ func NewAnalyzeCmd(log logr.Logger) *cobra.Command {
 	analyzeCommand.Flags().StringVar(&analyzeCmd.mavenSettingsFile, "maven-settings", "", "path to a custom maven settings file to use")
 	analyzeCommand.Flags().StringVarP(&analyzeCmd.mode, "mode", "m", string(provider.FullAnalysisMode), "analysis mode. Must be one of 'full' or 'source-only'")
 	analyzeCommand.Flags().BoolVar(&analyzeCmd.jsonOutput, "json-output", false, "create analysis and dependency output as json")
+	analyzeCommand.Flags().StringVar(&analyzeCmd.jaegerEndpoint, "jaeger-endpoint", "", "jaeger endpoint to collect traces")
 
 	return analyzeCommand
 }
@@ -334,6 +338,8 @@ func listOptionsFromLabels(sl []string, label string) {
 		fmt.Println("available target technologies:")
 	}
 	for _, tech := range newSl {
+		tech = strings.TrimSuffix(tech, "+")
+		tech = strings.TrimSuffix(tech, "-")
 		fmt.Println(tech)
 	}
 }
@@ -378,6 +384,9 @@ func (a *analyzeCommand) getConfigVolumes() (map[string]string, error) {
 		}
 		javaConfig.InitConfig[0].ProviderSpecificConfig["mavenSettingsFile"] = fmt.Sprintf("%s/%s", ConfigMountPath, "settings.xml")
 	}
+	if Settings.JvmMaxMem != "" {
+		javaConfig.InitConfig[0].ProviderSpecificConfig["jvmMaxMem"] = Settings.JvmMaxMem
+	}
 
 	provConfig := []provider.Config{
 		{
@@ -413,12 +422,26 @@ func (a *analyzeCommand) getConfigVolumes() (map[string]string, error) {
 	}
 	err = os.WriteFile(filepath.Join(tempDir, "settings.json"), jsonData, os.ModePerm)
 	if err != nil {
-		a.log.V(1).Error(err, "failed to write provider config", "dir", tempDir, "file", "settings.json")
+		a.log.V(1).Error(err,
+			"failed to write provider config", "dir", tempDir, "file", "settings.json")
 		return nil, err
 	}
-	return map[string]string{
+
+	vols := map[string]string{
 		tempDir: ConfigMountPath,
-	}, nil
+	}
+	// attempt to create a .m2 directory we can use to speed things a bit
+	// this will be shared between analyze and dep command containers
+	m2Dir, err := os.MkdirTemp("", "m2-repo-")
+	if err != nil {
+		a.log.V(1).Error(err, "failed to create m2 repo", "dir", m2Dir)
+	} else {
+		vols[m2Dir] = M2Dir
+		a.log.V(1).Info("created directory for maven repo", "dir", m2Dir)
+		a.tempDirs = append(a.tempDirs, m2Dir)
+	}
+
+	return vols, nil
 }
 
 func (a *analyzeCommand) getRulesVolumes() (map[string]string, error) {
@@ -541,6 +564,11 @@ func (a *analyzeCommand) RunAnalysis(ctx context.Context, xmlOutputDir string) e
 		fmt.Sprintf("--output-file=%s", AnalysisOutputMountPath),
 		fmt.Sprintf("--context-lines=%d", 100),
 	}
+	if a.jaegerEndpoint != "" {
+		args = append(args, "--enable-jaeger")
+		args = append(args, "--jaeger-endpoint")
+		args = append(args, a.jaegerEndpoint)
+	}
 	if !a.analyzeKnownLibraries {
 		args = append(args,
 			fmt.Sprintf("--dep-label-selector=(!%s=open-source)", provider.DepSourceLabel))

cmd/root.go

@@ -24,7 +24,6 @@ var noCleanup bool
 
 // rootCmd represents the base command when called without any subcommands
 var rootCmd = &cobra.Command{
-	// TODO: better descriptions
 	Short:        "A cli tool for analysis and transformation of applications",
 	Long:         ``,
 	SilenceUsage: true,

cmd/settings.go

@@ -22,6 +22,7 @@ type Config struct {
 	RootCommandName string `env:"CMD_NAME" default:"kantra"`
 	PodmanBinary    string `env:"PODMAN_BIN" default:"/usr/bin/podman"`
 	RunnerImage     string `env:"RUNNER_IMG" default:"quay.io/konveyor/kantra"`
+	JvmMaxMem       string `env:"JVM_MAX_MEM" default:""`
 }
 
 func (c *Config) Load() error {

go.mod

@@ -40,7 +40,7 @@ require (
 	github.com/bombsimon/logrusr/v3 v3.1.0
 	github.com/codingconcepts/env v0.0.0-20200821220118-a8fbf8d84482
 	github.com/inconshreveable/mousetrap v1.1.0 // indirect
-	github.com/konveyor/analyzer-lsp v0.3.0-alpha.1
+	github.com/konveyor/analyzer-lsp v0.3.0-beta.1.2
 	github.com/shopspring/decimal v1.3.1 // indirect
 	github.com/sirupsen/logrus v1.9.3
 	github.com/spf13/pflag v1.0.5 // indirect

go.sum

@@ -36,8 +36,8 @@ github.com/inconshreveable/mousetrap v1.1.0 h1:wN+x4NVGpMsO7ErUn/mUI3vEoE6Jt13X2
 github.com/inconshreveable/mousetrap v1.1.0/go.mod h1:vpF70FUmC8bwa3OWnCshd2FqLfsEA9PFc4w1p2J65bw=
 github.com/invopop/yaml v0.1.0 h1:YW3WGUoJEXYfzWBjn00zIlrw7brGVD0fUKRYDPAPhrc=
 github.com/invopop/yaml v0.1.0/go.mod h1:2XuRLgs/ouIrW3XNzuNj7J3Nvu/Dig5MXvbCEdiBN3Q=
-github.com/konveyor/analyzer-lsp v0.3.0-alpha.1 h1:RilOnB9E6+zDSDQs4vBVPMITuSD3o+6P0dkXWW2H3YQ=
-github.com/konveyor/analyzer-lsp v0.3.0-alpha.1/go.mod h1:Rv2WcWfVMEGEWqn0Fl4U4NcmJYPrmWdPtaFE9KDVVF8=
+github.com/konveyor/analyzer-lsp v0.3.0-beta.1.2 h1:yZh5rTZYq1XYzDei3/zwZOqE8XktftaETNycV3Q7ACE=
+github.com/konveyor/analyzer-lsp v0.3.0-beta.1.2/go.mod h1:zJCmIq08X0kPvtU8ZSmz+mZmQfBt4hdy9enoEy1AQw4=
 github.com/kr/pretty v0.1.0/go.mod h1:dAy3ld7l9f0ibDNOQOHHMYYIIbhfbHSm3C4ZsoJORNo=
 github.com/kr/pretty v0.2.0 h1:s5hAObm+yFO5uHYt5dYjxi2rXrsnmRpJx4OYvIWUaQs=
 github.com/kr/pretty v0.2.0/go.mod h1:ipq/a2n7PKx3OHsz4KJII5eveXtPO4qwEXGdVfWzfnI=

hack/README.md

@@ -8,4 +8,37 @@
 
 ## General run
 
-* `sample_jakartaee_duke_analyze.sh`:  Clone and analyze a jakartaee sample
+* `sample_jakartaee_duke_analyze.sh`:  Clone and analyze a jakartaee sample
+
+## Gather analysis traces 
+
+Full traces of the analysis process can be collected via Jaeger. 
+
+First, run the Jaeger container that will collect traces from analysis:
+
+```sh
+podman run -d --net=host --name jaeger -e COLLECTOR_ZIPKIN_HOST_PORT=:9411 jaegertracing/all-in-one:1.23
+```
+
+> Note that we are running Jaeger in `host` network so that analyzer container can communicate with it later. There are several services running in the container, it could happen that the ports they use are pre-occupied on your host. In that case, you will need to free them up manually.
+
+The Jaeger collector listens on `14268` while the UI listens on `16686` in the container.
+
+Now, we will run analysis by enabling the tracer that will send traces to our Jaeger collector. 
+
+To do that, pass `--jaeger-endpoint` flag with the value being the collector endpoint:
+
+```sh
+kantra analyze <YOUR_OPTIONS_HERE> --jaeger-enpoint 'http://172.17.0.1:14268/api/traces'
+```
+
+> Note that `172.17.0.1` is the IP address using which a Podman container can communicate with the host on Linux. It might be different for your system. Alternatively, you can create a network, set it as default and create your Jaeger instance in it to access it via a name instead of IP.
+
+When analysis finishes, download the traces from Jaeger:
+
+```sh
+curl -o traces.json http://localhost:16686/api/traces?service=analyzer-lsp
+```
+
+To view the results locally, open [http://localhost:16686/](http://localhost:16686/) in your browser
+